We introduce AudioLM, a framework for high-quality audio generation with long-term consistency. AudioLM maps the input audio to a sequence of discrete tokens and casts audio generation as a language modeling task in this representation space. We show how existing audio tokenizers provide different trade-offs between reconstruction quality and long-term structure, and we propose a hybrid tokenization scheme to achieve both objectives. Namely, we leverage the discretized activations of a masked language model pre-trained on audio to capture long-term structure and the discrete codes produced by a neural audio codec to achieve high-quality synthesis. By training on large corpora of raw audio waveforms, AudioLM learns to generate natural and coherent continuations given short prompts. When trained on speech, and without any transcript or annotation, AudioLM generates syntactically and semantically plausible speech continuations while also maintaining speaker identity and prosody for unseen speakers. Furthermore, we demonstrate how our approach extends beyond speech by generating coherent piano music continuations, despite being trained without any symbolic representation of music.
相關內容
This paper presents an overview and evaluation of some of the end-to-end ASR models on long-form audios. We study three categories of Automatic Speech Recognition(ASR) models based on their core architecture: (1) convolutional, (2) convolutional with squeeze-and-excitation and (3) convolutional models with attention. We selected one ASR model from each category and evaluated Word Error Rate, maximum audio length and real-time factor for each model on a variety of long audio benchmarks: Earnings-21 and 22, CORAAL, and TED-LIUM3. The model from the category of self-attention with local attention and global token has the best accuracy comparing to other architectures. We also compared models with CTC and RNNT decoders and showed that CTC-based models are more robust and efficient than RNNT on long form audio.
Data-driven approaches hold promise for audio captioning. However, the development of audio captioning methods can be biased due to the limited availability and quality of text-audio data. This paper proposes a SynthAC framework, which leverages recent advances in audio generative models and commonly available text corpus to create synthetic text-audio pairs, thereby enhancing text-audio representation. Specifically, the text-to-audio generation model, i.e., AudioLDM, is used to generate synthetic audio signals with captions from an image captioning dataset. Our SynthAC expands the availability of well-annotated captions from the text-vision domain to audio captioning, thus enhancing text-audio representation by learning relations within synthetic text-audio pairs. Experiments demonstrate that our SynthAC framework can benefit audio captioning models by incorporating well-annotated text corpus from the text-vision domain, offering a promising solution to the challenge caused by data scarcity. Furthermore, SynthAC can be easily adapted to various state-of-the-art methods, leading to substantial performance improvements.
We present DFormer, a novel RGB-D pretraining framework to learn transferable representations for RGB-D segmentation tasks. DFormer has two new key innovations: 1) Unlike previous works that aim to encode RGB features,DFormer comprises a sequence of RGB-D blocks, which are tailored for encoding both RGB and depth information through a novel building block design; 2) We pre-train the backbone using image-depth pairs from ImageNet-1K, and thus the DFormer is endowed with the capacity to encode RGB-D representations. It avoids the mismatched encoding of the 3D geometry relationships in depth maps by RGB pre-trained backbones, which widely lies in existing methods but has not been resolved. We fine-tune the pre-trained DFormer on two popular RGB-D tasks, i.e., RGB-D semantic segmentation and RGB-D salient object detection, with a lightweight decoder head. Experimental results show that our DFormer achieves new state-of-the-art performance on these two tasks with less than half of the computational cost of the current best methods on two RGB-D segmentation datasets and five RGB-D saliency datasets. Our code is available at: //github.com/VCIP-RGBD/DFormer.
With the rapid development of communication technologies and extended reality (XR), the services and applications of the Metaverse are gradually entering our lives. However, the current development of the Metaverse provides users with services that are homogeneous with the user experience that the Internet has brought in the past, making them more like an extension of the Internet. In addition, as a mobile application carrier for the Metaverse, it is also worth considering how vehicles with diverse onboard components can develop in synergy with the Metaverse. In this article, we focus on the core of the Metaverse, namely user experience, and provide a road map from Metaverse to Metaverse vehicles (Metavehicles). Specifically, we first elaborate on six features of the Metaverse from the perspective of user experience and propose a hierarchical framework for the Metaverse based on the evolutionary logic of the features. Under the guidance of this framework, we discuss the empowerment of onboard components of Metavehicles on the development of the Metaverse, and analyze the service experience that Metavehicles can bring to two types of users, namely drivers and passengers. Finally, considering the differentiated development levels of Metaverse and autonomous driving, we further establish a hierarchical framework for Metavehicles from three aspects (i.e., enhance Metaverse, enhance driving experience, and enhance entertainment experience), providing an evolutionary path for the development of Metavehicles.
In this paper, we introduce a large-scale and high-quality audio-visual speaker verification dataset, named VoxBlink. We propose an innovative and robust automatic audio-visual data mining pipeline to curate this dataset, which contains 1.45M utterances from 38K speakers. Due to the inherent nature of automated data collection, introducing noisy data is inevitable. Therefore, we also utilize a multi-modal purification step to generate a cleaner version of the VoxBlink, named VoxBlink-clean, comprising 18K identities and 1.02M utterances. In contrast to the VoxCeleb, the VoxBlink sources from short videos of ordinary users, and the covered scenarios can better align with real-life situations. To our best knowledge, the VoxBlink dataset is one of the largest publicly available speaker verification datasets. Leveraging the VoxCeleb and VoxBlink-clean datasets together, we employ diverse speaker verification models with multiple architectural backbones to conduct comprehensive evaluations on the VoxCeleb test sets. Experimental results indicate a substantial enhancement in performance,ranging from 12% to 30% relatively, across various backbone architectures upon incorporating the VoxBlink-clean into the training process. The details of the dataset can be found on //voxblink.github.io
Generative adversarial network (GAN)-based neural vocoders have been widely used in audio synthesis tasks due to their high generation quality, efficient inference, and small computation footprint. However, it is still challenging to train a universal vocoder which can generalize well to out-of-domain (OOD) scenarios, such as unseen speaking styles, non-speech vocalization, singing, and musical pieces. In this work, we propose SnakeGAN, a GAN-based universal vocoder, which can synthesize high-fidelity audio in various OOD scenarios. SnakeGAN takes a coarse-grained signal generated by a differentiable digital signal processing (DDSP) model as prior knowledge, aiming at recovering high-fidelity waveform from a Mel-spectrogram. We introduce periodic nonlinearities through the Snake activation function and anti-aliased representation into the generator, which further brings the desired inductive bias for audio synthesis and significantly improves the extrapolation capacity for universal vocoding in unseen scenarios. To validate the effectiveness of our proposed method, we train SnakeGAN with only speech data and evaluate its performance for various OOD distributions with both subjective and objective metrics. Experimental results show that SnakeGAN significantly outperforms the compared approaches and can generate high-fidelity audio samples including unseen speakers with unseen styles, singing voices, instrumental pieces, and nonverbal vocalization.
Many datasets have been designed to further the development of fake audio detection, such as datasets of the ASVspoof and ADD challenges. However, these datasets do not consider a situation that the emotion of the audio has been changed from one to another, while other information (e.g. speaker identity and content) remains the same. Changing the emotion of an audio can lead to semantic changes. Speech with tampered semantics may pose threats to people's lives. Therefore, this paper reports our progress in developing such an emotion fake audio detection dataset involving changing emotion state of the origin audio named EmoFake. The fake audio in EmoFake is generated by open source emotion voice conversion models. Furthermore, we proposed a method named Graph Attention networks using Deep Emotion embedding (GADE) for the detection of emotion fake audio. Some benchmark experiments are conducted on this dataset. The results show that our designed dataset poses a challenge to the fake audio detection model trained with the LA dataset of ASVspoof 2019. The proposed GADE shows good performance in the face of emotion fake audio.
Normalization is known to help the optimization of deep neural networks. Curiously, different architectures require specialized normalization methods. In this paper, we study what normalization is effective for Graph Neural Networks (GNNs). First, we adapt and evaluate the existing methods from other domains to GNNs. Faster convergence is achieved with InstanceNorm compared to BatchNorm and LayerNorm. We provide an explanation by showing that InstanceNorm serves as a preconditioner for GNNs, but such preconditioning effect is weaker with BatchNorm due to the heavy batch noise in graph datasets. Second, we show that the shift operation in InstanceNorm results in an expressiveness degradation of GNNs for highly regular graphs. We address this issue by proposing GraphNorm with a learnable shift. Empirically, GNNs with GraphNorm converge faster compared to GNNs using other normalization. GraphNorm also improves the generalization of GNNs, achieving better performance on graph classification benchmarks.
Collecting supporting evidence from large corpora of text (e.g., Wikipedia) is of great challenge for open-domain Question Answering (QA). Especially, for multi-hop open-domain QA, scattered evidence pieces are required to be gathered together to support the answer extraction. In this paper, we propose a new retrieval target, hop, to collect the hidden reasoning evidence from Wikipedia for complex question answering. Specifically, the hop in this paper is defined as the combination of a hyperlink and the corresponding outbound link document. The hyperlink is encoded as the mention embedding which models the structured knowledge of how the outbound link entity is mentioned in the textual context, and the corresponding outbound link document is encoded as the document embedding representing the unstructured knowledge within it. Accordingly, we build HopRetriever which retrieves hops over Wikipedia to answer complex questions. Experiments on the HotpotQA dataset demonstrate that HopRetriever outperforms previously published evidence retrieval methods by large margins. Moreover, our approach also yields quantifiable interpretations of the evidence collection process.
We construct targeted audio adversarial examples on automatic speech recognition. Given any audio waveform, we can produce another that is over 99.9% similar, but transcribes as any phrase we choose (at a rate of up to 50 characters per second). We apply our iterative optimization-based attack to Mozilla's implementation DeepSpeech end-to-end, and show it has a 100% success rate. The feasibility of this attack introduce a new domain to study adversarial examples.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191